CN114096194A

CN114096194A - Systems and methods for cognitive training and monitoring

Info

Publication number: CN114096194A
Application number: CN202080040282.1A
Authority: CN
Inventors: 耶尔·吉卢茨; 沙伊·格拉诺特; 安娜·伊左特奇夫
Original assignee: Aceral Co ltd
Current assignee: Aceral Co ltd
Priority date: 2019-05-30
Filing date: 2020-05-04
Publication date: 2022-02-25
Also published as: JP7311637B2; JP2022535799A; EP3975846A1; US20220230731A1; JP2023118913A; EP3975846A4; WO2020240534A1

Abstract

A system and method for analyzing user feedback in response to a cognitive training program, comprising: training at least one machine learning algorithm with a predefined data set to predict a training success rate, wherein the predefined data set includes previously received user feedback for users having known characteristics; receiving new user feedback; and determining a prediction of training success rate using at least one machine learning algorithm based on the received new user feedback.

Description

Systems and methods for cognitive training and monitoring

Technical Field

The present invention relates to cognitive training. More particularly, the present invention relates to systems and methods for monitoring and analyzing user feedback in response to cognitive training programs.

Background

People with cognitive problems, or those seeking improved cognitive skills, sometimes use cognitive training programs to improve their cognitive health and train their memory, similar to gym sports training. For example, a person may use a memory card, solve a crossword puzzle, or sit in front of a computer screen and perform various tasks directed to improving cognitive abilities (e.g., memory, computing, vocabulary, etc.).

A major problem with self-administered cognitive training programs (e.g., without any professional supervision) is user adherence to and/or participation in the training program. The time of use of a weekly, e.g. repetitive, program (where the user gradually loses interest in the program) may decrease and the training results decrease accordingly. For any exercise to be effective, people need to continue exercising for a long time. While most people typically begin training with high motivation to improve their cognitive abilities, ordinary users do not complete the training and over time they may stop or significantly reduce the amount of training sessions. Often, this may occur due to tedious training, difficulty in understanding how these activities correlate with daily needs, and/or due to lost interest. Thus, current training programs do not produce significant cognitive effects when self-administered.

SUMMARY

There is therefore provided, in accordance with some embodiments of the present invention, a method of analyzing user feedback in response to a cognitive training program, including: training, by a processor, at least one machine learning algorithm with a predefined data set to predict a training success rate, wherein the predefined data set may include previously received user feedback for users having known characteristics; receiving, by the processor, new user feedback; and determining, by the processor, a prediction of training success rate based on the received new user feedback using at least one machine learning algorithm. In some embodiments, at least one machine learning algorithm may be trained with reinforcement learning.

In some embodiments, the behavior pattern may be determined from user feedback. In some embodiments, the at least one machine learning algorithm may be implemented on a recurrent neural network with long and short term storage units. In some embodiments, a training churn rate (churn rate) may be predicted. In some embodiments, the received feedback may be monitored for at least one of timing, training session length, training session success rate, attention stability, freeze period, location, training platform, and number of breaks in the training session.

In some embodiments, the user feedback may be categorized to determine a user profile from a predefined profile list, wherein the prediction of the determined training success rate may also be based on the determined user profile. In some embodiments, the user profile may be determined based on at least one user characteristic selected from the group consisting of gender, age, education, location, language, occupation, current professional status, medical status, and marital status. In some embodiments, the user profile may be determined based on a cluster of received feedback and based on at least one user characteristic.

In some embodiments, a user may be monitored with at least one electroencephalogram (EEG) sensor, wherein a cognitive training program may be altered based on measured EEG signals. In some embodiments, the eye movement of the user may be monitored with at least one imager to determine the user's attention. In some embodiments, a behavior pattern may be determined from user feedback, and an alert may be issued when the determined behavior pattern exceeds a predefined threshold.

There is therefore provided, in accordance with some embodiments of the present invention, a system for cognitive analysis of user feedback in response to a cognitive training program, the system including: a database comprising a data set of previously received user feedback for users having known characteristics; and a processor coupled to the database and configured to: the method includes training at least one machine learning algorithm with the data set to predict a training success rate, receiving new user feedback, and determining a prediction of the training success rate using the at least one machine learning algorithm based on the received new user feedback. In some embodiments, at least one machine learning algorithm may be trained with reinforcement learning.

In some embodiments, the processor may determine the behavior pattern from user feedback. In some embodiments, the processor may classify the user feedback to determine a user profile from a predefined list of profiles, wherein the prediction of the training success rate using the at least one machine learning algorithm may also be based on the determined user profile. In some embodiments, the processor may predict the training churn rate using at least one machine learning algorithm. In some embodiments, the processor may monitor the received feedback for at least one of timing, training session length, training session success rate, attention stability, freeze period, and number of breaks in the training session.

In some embodiments, the at least one machine learning algorithm may be implemented on a recurrent neural network with long and short term storage units. In some embodiments, the user profile may be determined based on at least one user characteristic selected from the group consisting of gender, age, education, location, language, occupation, current occupation status, and marital status. In some embodiments, the user profile may be determined based on a cluster of received feedback. In some embodiments, at least one electroencephalogram (EEG) sensor may be coupled to the processor, wherein the processor may monitor the user with the at least one EEG sensor, and wherein the user profile may be determined based on the measured EEG signals.

In some embodiments, the at least one imager may be coupled to the processor, and wherein the processor may monitor eye movement of the user with the at least one imager. In some embodiments, the processor may determine a behavior pattern from the user feedback and issue an alert when the determined behavior pattern exceeds a predefined threshold.

There is therefore provided, in accordance with some embodiments of the present invention, a method of cognitive training, including: determining, by a processor, a behavior pattern from the received user feedback in response to the cognitive training program; and correcting, by the processor, the cognitive training program using at least one machine learning algorithm based on the determined behavior pattern to improve cognitive training. In some embodiments, at least one machine learning algorithm may be trained with previously received user feedback for users with known characteristics.

There is therefore provided, in accordance with some embodiments of the present invention, a method of analyzing user feedback in response to a cognitive training program, including: training, by a processor, at least one machine learning algorithm with a predefined data set to predict a training success rate, wherein the predefined data set may include previously received user feedback for users having known characteristics; and updating, by the processor, the training variables based on the prediction of the training success rate using the at least one machine learning algorithm. In some embodiments, new user feedback may be received, and the training variables may be re-updated based on the received new user feedback.

There is therefore provided, in accordance with some embodiments of the present invention, a method of analyzing user feedback in response to a cognitive training program, including: training, by a processor, at least one machine learning algorithm with a predefined data set to predict a training churn rate, wherein the predefined data set may include previously received user feedback for users having known characteristics; receiving, by the processor, new user feedback; and determining, by the processor, a prediction of training attrition rate using at least one machine learning algorithm based on the received new user feedback.

There is therefore provided, in accordance with some embodiments of the present invention, a method of analyzing user feedback in response to a cognitive training program, including: training, by a processor, at least one machine learning algorithm to predict cognitive decline (cognitive decline) using a predefined dataset, wherein the predefined dataset may include previously received user feedback for users having known characteristics; receiving, by the processor, new user feedback; and determining, by the processor, a prediction of cognitive decline using at least one machine learning algorithm based on the received new user feedback.

Brief Description of Drawings

The subject matter regarded as the invention is particularly pointed out and distinctly claimed in the concluding portion of the specification. The invention, however, both as to organization and method of operation, together with objects, features, and advantages thereof, may best be understood by reference to the following detailed description when read with the accompanying drawings in which:

FIG. 1 illustrates a block diagram of an example computing device, in accordance with some embodiments of the invention;

2A-2E illustrate block diagrams of systems for cognitive analysis of user feedback in response to a cognitive training program, in accordance with some embodiments of the invention;

FIG. 3 illustrates a block diagram of a system for cognitive data collection according to some embodiments of the invention;

4A-4B illustrate a flow diagram of a method for cognitive analysis of user feedback in response to a cognitive training program in accordance with some embodiments of the invention;

FIG. 5 illustrates a flow diagram of a method of analyzing user feedback to determine a training churn rate in response to a cognitive training program, according to some embodiments of the present invention; and

fig. 6 illustrates a flow diagram of a method of analyzing user feedback to determine cognitive decline in response to a cognitive training program, according to some embodiments of the invention.

It will be appreciated that for simplicity and clarity of illustration, elements shown in the figures have not necessarily been drawn to scale. For example, the dimensions of some of the elements may be exaggerated relative to other elements for clarity. Further, where considered appropriate, reference numerals may be repeated among the figures to indicate corresponding or analogous elements.

Detailed description of the invention

In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known methods, procedures, and components, modules, units, and/or circuits have not been described in detail so as not to obscure the present invention. Some features or elements described in relation to one embodiment may be combined with features or elements described in relation to other embodiments. For clarity, discussion of the same or similar features or elements may not be repeated.

Although embodiments of the invention are not limited in this regard, discussions utilizing terms such as, for example, "processing," "computing," "calculating," "determining," "establishing", "analyzing", "checking", or the like, may refer to operation(s) and/or process (es) of a computer, a computing platform, a computing system, or other electronic computing device, that manipulate and/or transform data represented as physical (e.g., electronic) quantities within the computer's registers and/or memories into other data similarly represented as physical quantities within the computer's registers and/or memories or other information non-transitory storage medium that may store instructions for performing operations and/or processes. Although embodiments of the present invention are not limited in this regard, the terms "plurality" and "a plurality" as used herein may include, for example, "multiple" or "two or more. The terms "plurality" or "a plurality" may be used throughout the specification to describe two or more components, devices, elements, units, parameters and the like. The term "group (set)" as used herein may include one or more items. Unless explicitly stated, the method embodiments described herein are not limited to a particular order or sequence. Additionally, some of the described method embodiments or some of their elements may occur or be performed synchronously, at the same point in time, or simultaneously.

Referring to FIG. 1, a schematic block diagram of an example computing device is shown, in accordance with some embodiments of the present invention. Computing device 100 may include a controller or processor 105 (e.g., a central processing unit processor (CPU), chip, or any suitable computing or computing device), an operating system 115, memory 120, executable code 125, storage 130, input devices 135 (e.g., a keyboard or touchscreen) and output devices 140 (e.g., a display), a communication unit 145 (e.g., a cellular transmitter or modem, a Wi-Fi communication unit, etc.) for communication with remote devices via a communication network, such as, for example, the internet. The controller 105 may be configured to execute program code to perform the operations described herein. The systems described herein may include one or more computing devices 100, for example, for use as the various devices and/or components shown in fig. 2A. For example, system 200 may be or may include computing device 100 or components thereof.

Operating system 115 may be or include any code segment (e.g., a code segment similar to executable code 125 described herein) designed and/or configured to perform tasks related to coordinating, scheduling, arbitrating, supervising, controlling, or otherwise managing the operation of computing device 100 (e.g., scheduling the execution of software programs or enabling software programs or other modules or units to communicate).

The memory 120 may be or include, for example, Random Access Memory (RAM), Read Only Memory (ROM), Dynamic RAM (DRAM), synchronous DRAM (SD-RAM), Double Data Rate (DDR) memory chips, flash memory, volatile memory, non-volatile memory, cache memory, buffers, short term memory units, long term memory units, or other suitable memory units or storage units. The memory 120 may be or may include a plurality of possibly different memory units. The memory 120 may be a computer or processor non-transitory readable medium, or a computer non-transitory storage medium, such as RAM.

Executable code 125 may be any executable code, such as an application, program, process, task, or script. Executable code 125 may be executed by controller 105 under the control of operating system 115. For example, executable code 125 may be a software application that performs the methods described further herein. Although a single item of executable code 125 is shown in FIG. 1 for clarity, a system according to embodiments of the invention may include multiple executable code segments similar to executable code 125 that may be stored in memory 120 and cause controller 105 to perform the methods described herein.

The storage 130 may be or include, for example, a hard disk drive, a Universal Serial Bus (USB) device, or other suitable removable and/or fixed storage unit. Furthermore, in some embodiments, some of the components shown in fig. 1 may be omitted. For example, memory 120 may be a non-volatile memory having the storage capacity of storage 130. Thus, although shown as a separate component, storage 130 may be embedded or included in memory 120.

The input device 135 may be or include a keyboard, a touch screen or pad, one or more sensors, or any other or additional suitable input device. Any suitable number of input devices 135 may be operatively connected to computing device 100. Output device 140 may include one or more displays or monitors and/or any other suitable output device. Any suitable number of output devices 140 may be operatively connected to computing device 100. As shown in

blocks

135 and 140, any suitable input/output (I/O) device may be connected to the computing device 100. For example, a wired or wireless Network Interface Card (NIC), a Universal Serial Bus (USB) device, or an external hard drive may be included in the input device 135 and/or the output device 140.

Embodiments of the invention may include an article such as a computer or processor non-transitory readable medium or a computer or processor non-transitory storage medium, such as, for example, a memory, disk drive, or USB flash memory that encodes, includes, or stores instructions (e.g., computer-executable instructions) that when executed by a processor or controller perform the methods disclosed herein. For example, the article may include a storage medium such as memory 120, computer-executable instructions such as executable code 125, and a controller such as controller 105. Such a non-transitory computer readable medium may be, for example, a memory, a disk drive, or a USB flash memory, encoding, including, or storing instructions, e.g., computer executable instructions, that when executed by a processor or controller, perform the methods disclosed herein. The storage medium may include, but is not limited to, any type of disk including semiconductor devices such as Read Only Memory (ROM) and/or Random Access Memory (RAM), flash memory, Electrically Erasable Programmable Read Only Memory (EEPROM), or any type of media suitable for storing electronic instructions, including programmable storage devices. For example, in some embodiments, memory 120 is a non-transitory machine-readable medium.

Systems according to embodiments of the invention may include components such as, but not limited to, multiple Central Processing Units (CPUs) or any other suitable multi-purpose or special purpose processor or controller (e.g., a controller similar to controller 105), multiple input units, multiple output units, multiple memory units, and multiple storage units. The system may additionally include other suitable hardware components and/or software components. In some embodiments, the system may include or may be, for example, a personal computer, desktop computer, laptop computer, workstation, server computer, network device, or any other suitable computing device.

According to some embodiments, systems and methods are provided for personalizing a computerized cognitive training program configured to support cognitive health, particularly memory function, of an adult user.

Referring now to fig. 2A, a block diagram of a system 200 for cognitive analysis of user feedback in response to a cognitive training program 210 is shown, in accordance with some embodiments. In fig. 2A, the direction of the arrows indicate the direction of information flow, and the dashed elements indicate software and/or algorithms.

The system 200 may include at least one processor 201 (such as the controller 105 shown in fig. 1), for example, a processor in a mobile device and/or PC on which the cognitive training program 210 may be implemented. The processor 201 may be coupled to a database 202 (such as the storage system 130 shown in fig. 1), the database 202 including a data set of previously received user feedback 203 for users having known characteristics. For example, a user with known characteristics such as age, gender, medical and/or mental condition may provide feedback 203 to the cognitive training program 210 (e.g., during an initial calibration phase described in detail herein).

During training, the processor 201 may collect information related to one or more of training time of day, training session length, platform used by the user (e.g., PC, tablet, smartphone), and/or training location (e.g., home or public), to improve the cognitive training program 210 for future training by the user. In some embodiments, the processor 201 may also gather information from the user feedback 203, such as user response time (e.g., in a game) and/or answer type (e.g., correct/wrong/missed answer) in different scenarios displayed to the user, and/or target location on the display, and/or input type (e.g., using a keyboard or touch screen), and/or number of times the user has rested. In addition to collecting information relating to answer types, the processor 201 may also collect information relating to at least one of: success rate, attention stability, lapse of attention, spatial attention, longest tie-ins (e.g., number of consecutive correct answers in a game), learning curve, sleep quality, and/or mood (e.g., based on questions in a training session or determined directly from a dedicated device such as a smart watch or other sensor). In some embodiments, the information collected by the processor 201 may be stored as user feedback 203 in the database 202.

According to some embodiments, the cognitive training program 210 may be modified based on cognitive functions recognized with a low success rate (e.g., in games, exercises, etc.), and/or a high standard deviation of response time (e.g., crossing some predefined threshold), and/or certain types of errors (e.g., location vs correctly recognized), etc. The trained cognitive functions may include memory components such as visual perception, combination of features and objects, organization of information, semantic networks, attention (critical to the memory process), and the like. The user's attention may be trained to focus, attention-directed, selective attention, visual spatial attention, sustained attention, perform attention and/or attention control (including, for example, distractive and frustrative), and the like. In some embodiments, the processor 201 may monitor the user feedback 203 to determine training attention based on the overall response time and recognize the lapse of attention (e.g., deviation in response time) and attention stability (e.g., magnitude of the overall standard deviation of response time).

The processor 201 may execute at least one machine learning algorithm 204 (e.g., with deep learning using a deep neural network) to train with a data set of user feedback 203 and predict a training success rate 205 for the user. The at least one machine learning algorithm 204 may be trained using data previously collected on the user while operating with predefined rules. In some embodiments, the at least one machine learning algorithm 204 may be trained with supervised training of computer networks using machine learning (e.g., with neural networks). Supervised training may include training with labeled data sets (e.g., such as training a human user with the system), or training under the supervision of a human operator who labels samples to teach the network. In some embodiments, once a predetermined amount of new data is collected, at least one machine learning algorithm 204 may be activated (e.g., semi-automatically) for a predetermined period of time for retraining of the algorithm. In some embodiments, the at least one machine learning algorithm 204 may be implemented on a Recurrent Neural Network (RNN), for example, with Long Short Term Memory (LSTM) elements. RNNs are a class of artificial neural networks in which connections between nodes form a directed graph along a time series. Unlike feed-forward neural networks, RNNs with LSTM structures may have feedback connections to process input sequences using their internal states (memory).

In some embodiments, the processor 201 may predict the training success rate 205 (e.g., using supervised learning), where similar users may be identified, for example, from continuously updated data sets of long-term users based on similar content (e.g., gender, location, age, education, etc.) and/or based on similar behavior (e.g., success with respect to their training history). Thus, in some embodiments, the training success rate 205 for a particular user may be calculated and/or predicted by the processor 201 based on the success of similar users for a particular training session.

In some embodiments, the cognitive training program 210 may receive input from the at least one machine learning algorithm 204 (e.g., directly and/or via the processor 201) including, for example, which exercises and/or levels and/or variables to use in order to receive the expected training success rate 205. In some embodiments, the processor 201 may continuously receive user feedback 203 with information of training progress according to provided instructions during and/or after training in order to compare training results to a predicted training success rate 205 in order to continuously improve at least one machine learning algorithm 204.

In some embodiments, the processor 201 may use at least one machine learning algorithm 204 (e.g., execute an algorithm to obtain results) to generate a recommendation to increase (or decrease) the training success rate 205. During training, the processor 201 may modify the cognitive training program 210 according to the calculated training success rate 205, e.g., if the calculated training success rate 205 is below a predefined threshold, the cognitive training program 205 may be more easily modified for the user. In some embodiments, the processor 201 may generate statistical data (e.g., charts) to be displayed to the user to reflect which variables affect their cognitive abilities, such as training time (e.g., time of day or day of week), sleep quality (e.g., best sleep time for maximum concentration), and/or whether there are differences in training sessions with respect to these variables. In some embodiments, at least some of the features may be determined by the processor 201 (e.g., success rate or training churn rate) to be performed by different machine learning algorithms.

Reference is now made to fig. 2B and 2C, which illustrate block diagrams of another

system

230 and 240, respectively, for determining a behavioral pattern 206 in response to a cognitive training program 210, in accordance with some embodiments. Some elements in fig. 2B and 2C may be the same as or similar to elements shown in fig. 2A (e.g., processor 201).

According to some embodiments, the processor 201 may determine a behavioral pattern 206 of the user (e.g., a pattern of behaviors and/or feedback, such as reaction times and/or correct answers that may reflect cognitive and/or motor abilities) from the user feedback 203, for example, using a machine learning algorithm 204. In some embodiments, the behavioral pattern 206 may be a cognitive behavioral pattern.

The processor 201 may utilize at least one machine learning algorithm 204 to determine a prediction of training success rate 205 based on the determined user behavior pattern 206 and/or based on user characteristics. In some embodiments, the processor 201 may issue an alert when the determined user behavior pattern 206 exceeds a predefined threshold. For example, the machine learning algorithm 204 may compare the user's behavior during training (e.g., from the user feedback 203) to an initial state (e.g., from the user profile 207) to determine whether the user's behavior pattern 206 meets or exceeds a predefined threshold (such as, for example, determining that the training success rate has decreased by 40%). In some embodiments, the information collected in the alert procedure may be fed back into the behavior pattern 206 algorithm of the user. For example, sliding window techniques may be used with supervised learning (e.g., RNN) or unsupervised learning by finding differences between previous windows. For both cases, a distance function between subsequences can be defined and used to calculate the distance to the previous window as input for several anomaly detection methods.

For example, the machine learning algorithm 204 may be configured to achieve a predefined training success rate 205 (e.g., 80%) for each user, where the training success rate 205 is measured, for example, during a training session to keep the user training for a considerable time (e.g., if a reduced success rate also indicates a reduction in training persistence), and/or when a new level of training is successfully completed (e.g., in a game), the training success rate 205 is measured. When a decrease in success rate 205 is identified, for example due to a decrease in training time, the machine learning algorithm 204 may be configured to achieve a lower training success rate 205, thereby maintaining training for the user.

In some embodiments, a set of actions for interacting directly with a user may be defined based on the behavior pattern 206 of a particular user to further improve the training success rate 205 and/or if the user's behavior changes significantly. The user feedback data 203 may be collected with reinforcement and/or supervised learning regarding the impact of different actions on the user's performance. Based on the collected user feedback data 203, the predefined rules and/or machine learning algorithms may apply training and/or parameters to be presented accordingly. For each such action (e.g., initiating a phone call with the user, providing educational material, etc.), the impact on the training success rate 205 can be measured to see which actions improve the training success rate 205. In some embodiments, once system 200 and/or system 230 and/or system 240 learn the responses to the actions, at least one machine learning algorithm 204 can predict for each user which actions may be required and at what times during training to apply them accordingly.

In some embodiments, the processor 201 may classify the user feedback 203 (and/or the processor 201 may instruct an algorithm to perform the classification), for example, to determine at least one user profile 207, including age, gender, etc., from a list of predetermined profiles (e.g., stored in the database 202); for example, the at least one profile may be determined by the at least one machine learning algorithm 204. In some embodiments, the determined at least one user profile may be used by at least one of the

systems

200, 230, 240, 260, and 270 shown in fig. 2A-2E. In some embodiments, the determined prediction of the user behavior pattern 206 may also be based on the determined at least one user profile 207. In some embodiments, the at least one user profile 207 may also or alternatively be determined based on at least one user characteristic, such as gender, age, education, location, language, occupation, current professional status, medical status, and/or marital status. The at least one user profile 207 may also or alternatively be determined based on a cluster of received feedback.

Referring now to fig. 2D, a block diagram of another system 260 for determining a training churn rate 264 in response to the cognitive training program 210 is shown, in accordance with some embodiments. Some elements in fig. 2D may be the same as or similar to elements shown in fig. 2A (e.g., processor 201).

According to some embodiments, the processor 201 may predict the training churn rate 264 using the machine learning algorithm 204. The user training attrition rate may be defined with different levels of ongoing engagement (e.g., based on training duration and/or training time of day).

In some embodiments, data of ongoing use and/or user engagement may be collected (e.g., if the user closes an account or stops training) in order to predict user training habits and/or expected training attrition rates 264 using machine learning algorithm 204. In some embodiments, machine learning algorithm 204 may receive as input data of other users previously identified as having reduced engagement and/or stopped training to compare together with the collected new training data of the users to predict expected training churn rate 264, e.g., also based on user profile 207.

According to some embodiments, the processor 201 may detect changes and/or anomalies in user behavior, for example, using unsupervised learning. The processor 201 may monitor user performance and define a profile for each user, each profile having expected behavior. In some embodiments, non-minor or significant deviations from expected behavior (e.g., predefined prior to training) detected from a new training session may be flagged to raise an alert, e.g., initiate contact with the user (e.g., call the user to try and understand the cause of the abnormality, or whether there is a medical problem or significant change, such as sadness, etc.).

In some embodiments, the measurement of user behavior may be performed at predetermined intervals, such as monthly, as an objective measurement or evaluation to see if there is any change in cognitive ability.

In some embodiments, the machine learning algorithm 204 may receive data for the user's behavior pattern 206 (shown in fig. 2C) as input to determine changes in behavior, such as determining the training churn rate 264.

Referring now to fig. 2E, a block diagram of another system 270 for determining cognitive decline 274 in response to the cognitive training program 210 is shown, in accordance with some embodiments. Some elements in fig. 2E may be the same as or similar to elements shown in fig. 2A (e.g., processor 201).

In some embodiments, the processor 201 may predict and/or detect cognitive decline 274 associated with Mild Cognitive Impairment (MCI) with the machine learning algorithm 204. MCI can lead to a significant and measurable decline in cognitive abilities including memory and mental skills (judgment, correct decision making, etc.). Persons with MCI are at increased risk of alzheimer's disease or other types of dementia.

In some embodiments, if users with different MCI phases are initially tagged, for example through diagnostics of an external medical facility, the at least one machine learning algorithm 204 may learn behavioral patterns of these users, for example, to later identify similar patterns of users that are not tagged at a certain phase of MCI. Accordingly, system 200 and/or system 270 can be used for prediction of MCI. In some embodiments, the multi-tag time series may be used with a few classes of prediction algorithms (for new users), such as with attention mechanisms or LSTM. An oversampling or generative countermeasure network (GAN) mechanism may be used to enhance the example set. In some embodiments, unsupervised detection may be used with algorithm-based clustering, such as Local Outlier Factors (LOFs), Kernel Density Estimates (KDEs), or K-Means, to identify the MCI level of a user.

In some embodiments, the machine learning algorithm 204 may receive as input data of other users previously identified as cognitive decline (e.g., with MCI or dementia) for comparison with the collected new training data of the user, e.g., to predict an expected cognitive decline 274, e.g., may also be based on the user profile 207.

In some embodiments, the machine learning algorithm 204 may receive as input data for a behavioral pattern 206 (shown in fig. 2C) of the user in order to determine changes in behavior, for example, to determine cognitive decline 274.

Referring now to fig. 3, fig. 3 illustrates a block diagram of a system 300 for cognitive data collection, according to some embodiments. In some embodiments, the system 300 may also include some or all of the elements of the system 200 (such as the processor 201 and the database 202) to which the elements of the system 300 are added to collect cognitive data from the user 30.

In some embodiments, the system 300 may comprise at least one electroencephalogram (EEG) sensor 301 coupled to the processor 201 to measure EEG signals, the processor 201 being configured to monitor cognitive signals of the user 30 with the at least one EEG sensor 301. In some embodiments, the user profile 207 may also be determined based on measured EEG signals. For example, the user 30 may wear a headset with at least one EEG sensor 301 to collect measurements on brain waves and specific activities using commercial EEG channels (1-16) or clinical electroencephalogram channels (16-64) depending on private use or clinician's license, respectively. In some embodiments, the determined training success rate 205 may be refined based on data collected by the at least one EEG sensor 301.

In some embodiments, the EEG sensors 301 may be similarly used to provide neurofeedback, following different brain waves (e.g., α, β, θ) and the relationships between them, to find correlations between brain waves, for example, in response to the cognitive training program 210. In some embodiments, the measured signals and specific responses to the measured waves and/or wave relationship thresholds may be integrated into a training session (e.g., into a game). In some embodiments, training may also include options for dual tasks with neurofeedback based on brain wave activity (e.g., based on alpha levels or theta or beta levels) and predetermined levels and/or measures to maintain while training.

In some embodiments, the system 300 may include, for example, at least one imager 302 coupled to the processor 201, the processor 201 being configured to monitor eye movements of the user 30 for content displayed on the display 310 with the at least one imager 302, and thus determine the concentration and/or attention of the user 30 during training. In some embodiments, the determined training success rate 205 may be refined, for example, based on data collected by the at least one imager 302.

In some embodiments, the at least one imager 302 may track eye movement and/or pupil size with a camera of a computerized device (such as computing device 100 shown in fig. 1), e.g., a tablet, smartphone, or the like, or through a clinical eye tracker. Eye tracking data may be collected during training and with respect to the state presented in the training display 310. In some embodiments, the processor 201 may analyze the collected eye tracking data to identify saccades, fixations, pupil sizes, etc. associated with content presented on the display 310 to determine attention quality, attention measurements, and memory measurements.

In some embodiments, at least some training sessions may be performed in a virtual reality environment. For example, a single wearable device (e.g., a headset) may include EEG sensors 301 and/or imager 302 and/or virtual reality imaging displayed with the headset to combine biofeedback with pulse rate and perspiration monitoring.

In some embodiments, the processor 201 may analyze data collected from external devices (such as EEG sensors 301 and/or imagers 302) as well as other external devices used by the user (such as activity trackers, smartwatches, smartphones, clinical data, and test results) to improve the cognitive training program 210 and increase the training success rate 205 accordingly. The additional collected data may relate to sleep quality, daily activity, location (e.g., using GPS data), stability (e.g., hand stability when holding the device), and/or emotional state (e.g., based on speech and speech recognition, and/or based on nutrition, medication, etc.). In some embodiments, the collected data may be computed using at least one machine learning algorithm 204, for example, to provide more accurate personalized training, personal recommendations, and/or cognitive tagging.

According to some embodiments, the processor 201 may calculate a training success rate 205 and/or a general training schedule based on the correct answer percentage and/or response time and based on spatial attention detection, for example, to measure how the user distracts in the surrounding space. Thus, in some embodiments, training may include targets in different regions of the display 310, and the received responses between these regions (e.g., registered as user feedback 203) may be compared, e.g., to create a spatial attention map and locate regions that are "ignored" (within the display region). These regions may be labeled and trained to increase the spatial attention of the user.

Referring now to fig. 4A, a flow diagram of a method of analyzing user feedback 203 in response to a cognitive training program 210 is shown, in accordance with some embodiments.

At least one machine learning algorithm 204 may be trained (e.g., by processor 201) at step 401 using a predefined data set to predict a training success rate 205, where the predefined data set may include previously received user feedback for users having known characteristics. New user feedback may be received at step 402, and a prediction of the training success rate 205 may be determined (e.g., by the processor 201) using at least one machine learning algorithm 204 based on the received new user feedback at step 403. In some embodiments, the at least one machine learning algorithm 204 may be trained with reinforcement learning. In some embodiments, a transfer learning algorithm may be used to build a predictive model of a person with cognitive decline using data from healthy persons.

Referring now to fig. 4B, a flow diagram of a method of analyzing user feedback 203 in response to a cognitive training program 210 is shown, in accordance with some embodiments. In some embodiments, at step 404, the at least one machine learning algorithm 204 may be trained (e.g., by the processor 201) with a predefined data set to predict the training success rate 205, wherein the predefined data set may include previously received user feedback for users having known characteristics. The training set and/or training variables may be determined (e.g., using a predetermined threshold) to satisfy the prediction of training success rate 205, where the training variables may be updated at step 405 based on the prediction of training success rate using the at least one machine learning algorithm 204. When new user feedback is received at step 406, the training variables may be updated again at step 405.

According to some embodiments, the at least one machine learning algorithm 204 may be trained (e.g., by the processor 201) with a predefined set of data, e.g., to determine the behavioral pattern 206 of the user, wherein the predefined set of data may include previously received user feedback for users having known characteristics and previously calculated behavioral patterns of other users. In some embodiments, new user feedback may be received, and a comparison of the behavioral pattern 206 of the user with the newly received data may be performed (e.g., by the processor 201) utilizing at least one machine learning algorithm 204 to identify anomalies in the behavioral pattern 206.

According to some embodiments, the at least one machine learning algorithm 204 may be trained (e.g., by the processor 201) with a predefined set of data, e.g., to determine the behavioral pattern 206 of the user, wherein the predefined set of data may include previously received user feedback for users having known characteristics and previously calculated behavioral patterns of other users. The at least one machine learning algorithm 204 may be trained to predict the behavior of the user in other situations accordingly, such as predicting the behavior in stress conditions for a particular user.

Referring now to fig. 5, a flow diagram of a method of analyzing user feedback 203 to determine a training churn rate 264 in response to a cognitive training program 210 is shown, according to some embodiments.

In some embodiments, at least one machine learning algorithm 204 may be trained (e.g., by processor 201) at step 501 with a predefined data set to predict training churn rate 264, where the predefined data set may include previously received user feedback for users with known characteristics (e.g., user feedback for number of training sessions, training frequency, training time/date, etc.). New user feedback may be received at step 502, and a prediction of training attrition rate 264 may be determined at step 503 (e.g., by processor 201) utilizing at least one machine learning algorithm 204 based on the received new user feedback.

Referring now to fig. 6, a flow diagram of a method of analyzing user feedback 203 to determine cognitive decline 274 in response to a cognitive training program 210 is shown, in accordance with some embodiments.

In some embodiments, the at least one machine learning algorithm 204 may be trained (e.g., by the processor 201) at step 601 with a predefined set of data to label possible cognitive decline and/or predict cognitive decline 274, wherein the predefined set of data may include previously received user feedback for users having known characteristics. The predefined data set may include previously calculated patterns characterizing different cognitive deterioration states, for example, based on previously received user feedback for users having known characteristics including cognitive clinical diagnosis. New user feedback may be received at step 602, and a prediction of cognitive decline 274 may be determined at step 603 (e.g., by processor 201) utilizing at least one machine learning algorithm 204 based on the received new user feedback.

While certain features of the invention have been illustrated and described herein, many modifications, substitutions, changes, and equivalents will now occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.

Various embodiments have been proposed. Of course, each of these embodiments may include features of other embodiments presented, and embodiments not specifically described may include various features described herein.

Claims

1. A method of analyzing user feedback in response to a cognitive training program, the method comprising:

training, by a processor, at least one machine learning algorithm with a predefined data set to predict a training success rate, wherein the predefined data set includes previously received user feedback for users having known characteristics;

receiving, by the processor, new user feedback; and

determining, by the processor, a prediction of the training success rate using the at least one machine learning algorithm based on the received new user feedback,

wherein the at least one machine learning algorithm is trained using reinforcement learning.

2. The method of claim 1, further comprising determining, by the processor, a behavior pattern from the user feedback.

3. The method of claim 1, wherein the at least one machine learning algorithm is implemented on a recurrent neural network with long and short term storage.

4. The method of claim 1, further comprising predicting, by the processor, a training churn rate.

5. The method of claim 1, further comprising monitoring, by the processor, the received feedback for at least one of timing, training session length, training session success rate, attention stability, freeze period, location, training platform, and number of breaks in a training session.

6. The method of claim 1, further comprising classifying, by the processor, the user feedback to determine a user profile from a predefined list of profiles, wherein the prediction of the determined training success rate is also based on the determined user profile.

7. The method of claim 6, wherein the user profile is further determined based on at least one user characteristic selected from the group consisting of gender, age, education, location, language, occupation, current occupation status, medical status, and marital status.

8. The method of claim 7, wherein the user profile is further determined based on a cluster of received feedback and based on at least one user characteristic.

9. The method recited in claim 1, further comprising monitoring, by the processor, a user with at least one electroencephalography (EEG) sensor, wherein the cognitive training program changes based on measured EEG signals.

10. The method of claim 1, further comprising monitoring, by the processor, eye movement of the user with at least one imager to determine the user's attention.

11. The method of claim 1, further comprising:

determining, by the processor, a behavior pattern from the user feedback; and

issuing an alert by the processor when the determined behavior pattern exceeds a predefined threshold.

12. A system for cognitive analysis of user feedback in response to a cognitive training program, the system comprising:

a database comprising a data set of previously received user feedback for users having known characteristics; and

a processor coupled to the database and configured to:

training at least one machine learning algorithm using the data set to predict a training success rate;

receiving new user feedback; and

determining a prediction of the training success rate using the at least one machine learning algorithm based on the received new user feedback,

13. The system of claim 12, wherein the processor is further configured to determine a behavior pattern from the user feedback.

14. The system of claim 12, wherein the processor is further configured to classify the user feedback to determine a user profile from a predefined list of profiles, wherein the prediction of training success rate using the at least one machine learning algorithm is also based on the determined user profile.

15. The system of claim 12, wherein the at least one machine learning algorithm is implemented on a recurrent neural network with long and short term storage.

16. The system of claim 12, wherein the processor is further configured to utilize the at least one machine learning algorithm to predict a training churn rate.

17. The system of claim 12, wherein the processor is further configured to monitor the received feedback for at least one of timing, training session length, training session success rate, attention stability, freeze period, and number of breaks in a training session.

18. The system of claim 12, wherein the user profile is further determined based on at least one user characteristic selected from the group consisting of gender, age, education, location, language, occupation, current occupation status, and marital status.

19. The system of claim 12, wherein the user profile is further determined based on a cluster of received feedback.

20. The system recited in claim 12, further comprising at least one electroencephalography (EEG) sensor coupled to the processor, wherein the processor is further configured to monitor a user with the at least one EEG sensor, and wherein the user profile is determined based on measured EEG signals.

21. The system of claim 12, further comprising at least one imager coupled to the processor, and wherein the processor is further configured to monitor eye movement of a user with the at least one imager.

22. The system of claim 12, wherein the processor is further configured to determine a behavior pattern from the user feedback, and issue an alert when the determined behavior pattern exceeds a predefined threshold.

23. A method of cognitive training, the method comprising:

determining, by the processor, a behavior pattern from the received user feedback in response to the cognitive training procedure; and

correcting, by the processor, the cognitive training program with at least one machine learning algorithm based on the determined behavior pattern to improve the cognitive training,

wherein the at least one machine learning algorithm is trained with previously received user feedback for users having known characteristics.

24. A method of analyzing user feedback in response to a cognitive training program, the method comprising:

training, by a processor, at least one machine learning algorithm with a predefined data set to predict a training success rate, wherein the predefined data set includes previously received user feedback for users having known characteristics; and

updating, by the processor, training variables according to the prediction of the training success rate using the at least one machine learning algorithm.

25. The method of claim 24, further comprising:

receiving, by the processor, new user feedback; and

re-updating, by the processor, the training variables based on the received new user feedback.

26. A method of analyzing user feedback in response to a cognitive training program, the method comprising:

training, by a processor, at least one machine learning algorithm with a predefined data set to predict a training churn rate, wherein the predefined data set includes previously received user feedback for users having known characteristics;

receiving, by the processor, new user feedback; and

determining, by the processor, a prediction of the training churn rate using the at least one machine learning algorithm based on the received new user feedback.

27. A method of analyzing user feedback in response to a cognitive training program, the method comprising:

training, by a processor, at least one machine learning algorithm with a predefined data set to predict cognitive decline, wherein the predefined data set includes previously received user feedback for users having known characteristics;

receiving, by the processor, new user feedback; and

determining, by the processor, a prediction of the cognitive decline with the at least one machine learning algorithm based on the received new user feedback.